Download 1 - EcoMod

DRAFT
How Can Public Spending Help You Grow?
An Empirical Analysis for Developing Countries1
Nihal Bayraktar
Penn State University and World Bank
and
Blanca Moreno-Dodson
World Bank
December 16, 2009
Abstract: Many endogenous growth models introduce public spending or its components as determinants of
economic growth. Even though many empirical studies indicate that both level and composition of public
spending are significant for economic growth, the results, however, are still mixed. We try to understand
the importance of sample selection to explain these conflicting results. Public spending can be a significant
determinant of growth for countries that are capable of using expenditures for productive purposes. We
investigate a set of fast-growing developing countries versus a mix of developing countries with different
growth patterns. In the regression specifications, we include different components of public expenditure
and fiscal revenues, always considering the overall government budget constraint.
Total public spending is disaggregated using a definition based on Bleaney, Kneller, and Glemmell (2001),
and Kneller, Bleaney and Gemmell (1998), which classifies public spending as productive versus
unproductive components (a priori). The empirical analysis shows that the link between public spending,
especially its productive components, and growth is strong only for the fast-growing group. The most
important factor that affects the magnitude of the influence of public spending on growth is the priority
given to productive over other spending. In addition, macroeconomic stability, openness, and private sector
investment are also significant in the fast-growing group, which points out to the existence of
complementarities between private and public sector spending.
JEL codes: H50, O11, O23
Key words: Public spending, productive expenditure, government budget, economic growth
1
The authors would like to thank Pierre Richard Agénor, Professor at the University of Manchester and
Center for Growth and Business Cycles Research; Vito Tanzi, former IMF Fiscal Affairs Department
Director; and Gilles Nancy, Professor at the University of Aix-Marseille II, for their helpful comments.
1
EXECUTIVE SUMMARY
Even though, in many studies, it has been shown that total public spending and some of
its components are significant for economic growth, the results are still mixed. We try to
understand the importance of country sample selection to explain these conflicting
results. We conclude that public spending can be a significant determinant of growth for
countries that are capable of using expenditures for productive purposes. As a follow up
to a previous study2, this paper investigates empirically how the impact of public
spending on growth varies when countries are classified according to their overall growth
performances. The analysis compares fast-growing versus a mix of countries with
different growth patterns. Seven countries are included in each group and the dataset
covers the period of 1970-2006. In the regression specifications, we include different
components of public expenditure and fiscal revenues, always applying the overall
government budget constraint. A priori, while the size of the government does not appear
to be much different (on average) in these two groups, the composition of public
expenditures varies significantly. The share of productive public spending is much larger
for the fast-growing countries in our dataset, while the share of unproductive components
of public spending is higher in the comparison group. The empirical analysis based on
OLS and dynamic GMM techniques for panel data shows that the link between public
spending, especially its productive component, and growth, after controlling for other
macroeconomic and private sector variables, and taking into account country initial
conditions, is both economically and statistically strong only for the fast-growing group.
When all countries are combined in the same regression analysis, the link between
government spending and growth gets much weaker on average and even turns negative
in some cases. However, when group effects in the combined set are controlled for,
through interactive dummies, the stronger link between public spending and growth in
the fast-growing group is clearly confirmed. A possible nonlinear relationship between
growth and public spending is also investigated but no statistical significance is found for
it. Our study shows that the most important factor that affects the degree of influence of
public spending on growth is the priority given to productive over other spending. In
addition, macroeconomic stability, openness, and private sector investment are also
significant in the fast-growing country group, which points out to the existence of
complementarities between private and public sector spending. A final implication of the
analysis is that differences in empirical findings of the previous literature linking public
spending and economic growth may be explained by the selection of countries included.
2
Moreno-Dodson (2008) shows that the volume of total public spending as well as its composition are
relevant in explaining economic growth for a set of fast-growing developing countries during 1970-2004.
2
I. INTRODUCTION
The importance of public spending and its components on economic growth has been
extensively studied in the literature. Within an endogenous growth framework, in 1990
Barro introduced public services in the production function. Many empirical studies have
followed this seminal paper to investigate the possible link between different components
of government spending and growth, using many different econometric techniques,
empirical settings, and samples of countries. Results presented in the literature are mixed.
Even though most studies support the substantial positive link between some components
of public spending and growth, there is still no agreement on which categories of
spending promote growth. The introduction of advanced econometric techniques and new
variables in the empirical specifications could not solve the problem.
One possible explanation for the mixed results obtained in the literature is sample
selection. What we expect is that public spending can improve growth performance of
countries only if they are able to use these expenditures productively. In light of this
expectation, we raise the following questions: (1) Are there any obvious differences
between fast-growing countries and others, in terms of the level of public spending, its
components, and their link to growth? (2) Is the link between public spending and
growth stronger for countries experiencing higher and sustained growth rates? (3) How
different are public spending and its link to economic growth in other developing
countries where growth records are less impressive? (4) Why do we observe differences
in the response of growth to public spending among countries with a similar level of
government spending? (5) What is the role of the composition of public spending on the
growth performance of countries?
Given that many governments in developing countries are considering increasing public
spending to provide a short-term economic stimulus in the midst of the current global
economic crisis, we believe that the questions above are more relevant than ever. Indeed,
they have important implications for changes in the composition of public expenditure, to
the extent that different allocations may involve dynamic tradeoffs in their short- and
medium-term impact on growth.
In addition to these questions, a possible nonlinear link between public spending and
growth (also studied in the previous paper) is analyzed. It is important to see whether the
link between different components of public expenditure and growth is statistically
significantly positive or not. But the shape of the function is also essential for policy
implications. For example, if the link between public spending and growth is positive and
concave, higher and higher public spending may have less and less impact on growth. On
the other hand, if the link is positive and convex, we expect higher public spending to
lead to an even relatively larger effect on growth. In order to capture possible nonlinear
effects, the square terms of public spending are introduced in the empirical model in a
dynamic growth context.3
3
Another type of nonlinearities associated with the effect of government spending on growth relates to
threshold effects. In particular, there is growing evidence that spending on infrastructure may be subject to
3
The analysis presented in this paper is applied to a sample including seven fast-growing
developing countries (the same group considered in Moreno-Dodson [2008]) 4 and seven
other developing countries whose growth rate patterns have been somehow less stable
and consistent during the period of analysis. The first set contains Korea, Singapore,
Malaysia, Thailand, Indonesia, Botswana, and Mauritius, which have been among the top
performers in the world in terms of GDP per capita growth during the period between
1960 and 2006.5 The second set includes Chile, Costa Rica, Mexico, Philippines,
Turkey, Uruguay and Venezuela, and is taken as a comparison group to enhance and
validate the econometric estimates of Moreno-Dodson (2008) and to further examine the
influence of public expenditures on growth.6 The paper presents a comparative analysis
of the countries and panel regressions conducted using OLS and dynamic GMM
techniques. Since we use annual data, the focus of the OLS results is on the short-run
analysis, while the GMM results focus on a dynamic, multi-year framework.
The paper is structured as follows. Section II presents the literature review. Section III
presents the data and provides relevant facts and information about the two groups of
countries during the period of analysis. Section IV describes the empirical methodology,
function specification, and variables selected. Section V is dedicated to the results
obtained with the panel regression analysis. Finally, Section VI draws policy implications
and concludes.
II. LITERATURE REVIEW
Despite the fact that the link between public expenditure and economic growth has been
investigated extensively in the literature, robust conclusions have been difficult to
establish. Even though, in recent studies, there is some convergence in terms of the
significance of public spending on growth, the results still change from country to
country, or from sample to sample, and can be a function of many different factors.
Some of the early empirical studies suggest that the link between public expenditure and
growth is positive. In his very influential paper, Barro (1990) extends the endogenous
growth framework including tax-financed government services. He concludes that
“critical mass” or “network” effects, which imply that its impact on growth becomes significant (or is
magnified) beyond a certain level; see for instance Pushak, Tiongson, and Varoudakis (2007) and
Kellenberg (2009). However, this type of nonlinearities is mostly associated with the stock of public assets,
rather than spending flows, as we discuss here.
4
Moreno-Dodson (2008) empirically investigates the impact of public spending and its components on
economic growth, focusing on a sample of seven fast-growing developing countries. She finds that some
components of public expenditure, particularly those considered “productive,” can significantly explain
economic growth. Despite the fact that there are some differences at the country level, the results are
consistent across different econometric techniques used to estimate the statistical significance of public
spending items.
5
All the countries selected in the sample have sustained GDP per capita growth rates of at least 3% (by
decade average) during 1960-2006.
6
It should be noted that the availability of data for the period of analysis has played an important role in
selecting these countries.
4
government expenditure is positively linked to economic growth when the share of
government expenditure (and consequently the tax rate) is low, but then turns negative
due to increasing inefficiencies as the share of expenditure increases (related to the
disincentive effect of higher tax rates on private capital accumulation), indicating a
nonlinear relationship between government expenditure and growth.7 In a similar setting,
Barro and Sala-i-Martin (1992) show that if social returns on public investment are larger
than private returns, public investment can increase economic growth. Barro (1991) tests
empirically the link between government expenditure and growth, using cross-country
analysis and including 98 developing countries for the period of 1960-1985. The study
finds that public consumption is negatively correlated with growth, while public
investment does not have a significant impact on economic development.
Studies including both developed and developing countries also present similar results.
Grossman (1990), using a sample consisting of 48 developed and developing countries,
shows that government spending has both positive and negative impacts on growth; the
positive one works through higher productivity and the negative one is caused by
inefficient provision and distortionary effects of public taxation. However, he concludes
that the positive influence dominates. There are also early studies indicating the opposite.
Levine and Renelt (1992) also show that taking into account the components of
government spending can make a difference. In their paper, they separate government
spending in two broad categories, consumption expenditure and investment outlays. For
119 developed and developing countries during the period of 1974 to 1989, they find a
negative relationship between government consumption and growth, but a clear positive
link between public investment and growth. Empirical studies also show that not only the
level of public spending but also its composition matters for economic growth. An
empirical study of Easterly and Rebelo (1993), using a sample similar to the one used by
Barro (1991), finds that public investment in communication and transportation as well as
general government investment promotes economic growth. Glomm and Ravikumar
(1997) review some of the papers studying the influence of productive government
expenditures on long-run growth and conclude that government expenditures on health,
education and infrastructure have large impacts on growth. Similarly, the results
presented by Kneller, Bleaney, and Gemmell (1999) and Bleaney, Gemmell, and Kneller
(2001), where 22 developed countries are included, support Barro (1990): productive
expenditure is good for growth, but distortionary taxes lower its impact.
When we focus on most recent studies, conflicting results still continue to be found. One
of the recent papers by Schaltegger and Torgler (2006) suggests that large public
expenditure lowers growth for high-income countries. Folster and Henrekson (2001)
suggest that the more the econometric problems that are addressed, the more robust the
link between government size and economic growth gets, while Agell, Ohlsson, and
Thoursie (2006) object to this finding, indicating that there is no robust relationship
between growth and the share of government expenditure. In his paper, Park (2006) tests
whether the combination of productive public investment and lower taxes increases
growth and whether current government consumption and higher taxes lower it. He
7
See Slemrod (1995) for the literature review on the link between government expenditure, taxes, and
growth.
5
cannot find any robust empirical results using a set of countries combining both
developed and developing countries. Gupta, Clements, Baldacci, and Mulas-Granados
(2002) show that government expenditure, especially its capital component, has a
positive impact on growth for developing countries when it is combined with a lower
budget deficit. Similarly, Wahab (2004) and Colombier (2009), focusing on OECD
countries, and Ang (2009), studying the case of Malaysia all support the significance of
public capital expenditure for growth.
In an empirical setting similar to our setting, Ghosh and Gregoriou (2008) show that the
current component of spending has a positive impact on growth, while the capital
component influences it negatively for a group of 15 developing countries. In a paper
where a similar empirical setting and econometric methodologies are used, Bleaney,
Gemmell, and Kneller (2001) show that the productive component of public spending,
including capital expenditures, is significant for growth in OECD countries.
While most empirical studies in the literature use a heterogeneous sample of countries to
investigate the relationship between government spending and growth, Moreno-Dodson
(2008) includes only fast-growing developing countries and shows that the link between
total public spending and growth is overall positive with some components of public
spending being particularly significant in affecting growth. For this group of countries,
unproductive components of public expenditure are less effective8—or even have a
negative impact on growth—while the productive component of public spending is
statistically significantly. 9
Our paper extends this initial empirical study in a way to answer the question of whether
the findings presented in that paper are sensitive to country sample selection bias. We try
to accomplish this goal by extending the initial data set and including a mix of countries
with different growth performances. We also try to identify, in a dynamic setting,
whether the links among government spending, its components, and growth are linear or
not when all countries are either investigated or combined separately.
Similarly, the empirical specification introduced in the paper includes the government
budget constraint to avoid biases associated with incomplete specification ignoring
financing options of governments and budget balance, in line with other recent papers in
the literature.10
8
Defined a priori using the definition by Bleaney, Gemmel , and Kneller (2001) and and Kneller, Bleaney
and Gemmell (1998).
9
Similarly, Rogers (2008) shows that the impact of public schooling expenditures on economic growth is
significantly higher in countries that are considered to use schooling productively.
10
For example, Bose, Haque, and Osborn (2007) introduce government financing variables (government
budget surplus/deficit and tax revenue) in a study where they focus on a panel of 30 developing countries
over the 1970s and 1980s. They find that while the capital component of government expenditure,
especially education expenditure, is positively linked to growth, the current component does not have any
significant impact on economic growth. Similarly, Ghosh and Gregoriou (2008) and Benos (2009) take
6
Another element introduced in the paper is the test of non-linearities in the empirical
specification with the government budget constraint. There are many other papers
focusing on a possible nonlinearity between government spending and growth. In
addition to Barro (1990 and 1991), as indicated above, Grossman (1988) examines a
nonlinear relationship between growth in the size of government and overall growth in
the economy. The results for the United States indicate that this relationship is explained
better with a nonlinear model. When compared with previous papers, the main difference
of our nonlinear analysis is that we consider a dynamic setting where the overall budget
constraint is also taken into account.
III. DATA AND COMPARATIVE ANALYSIS
Data
As indicated in the introduction and the previous section, two sets of countries are
included in the analysis. The first group consists of the seven fast-growing developing
countries that were included in Moreno-Dodson (2008).11 The regression period is 19702006.12 The main data source is the Government Financial Statistics (GFS) of the
International Monetary Fund (IMF).13
Comparative Analysis
This section investigates some country facts and findings, and compares different country
characteristics on public expenditure and growth that may be helpful when interpreting
the subsequent econometric results.
First, when the growth rates of GDP per capita in real terms are compared for these two
groups of countries, it can be seen that the first set, by definition, have grown much faster
on average between 1970 and 2005 (see Tables 1 and 2). While this group has almost 5
percent growth rate on average, the second group has a mean growth rate of 1.6 percent.
As can be seen in Figure 1, the first group outperforms the second set of countries
throughout the period of 1970 to 2005 except during the Asian financial crises of 1997-
account of the revenue side of the government budget constraint considering tax and non-tax revenues, and
also the government budget balance. Both papers use a GMM technique for panel datasets. But, they find
conflicting results. Benos (2009) show that a reallocation of the components of government spending,
especially toward infrastructure and human capital, can enhance growth using 14 European Union
countries, while Ghosh and Gregoriou (2008) indicate current (capital) spending has a positive (negative)
impact on the growth rate.
11
The list of countries in the fast-growing set and the comparison set is given in the introduction section.
The time period may change slightly from one country to another depending on data availability at the
country level.
13
Details on variables and data sources are presented in the Appendix to this paper as well as in MorenoDodson (2008).
12
7
98, and the years 2004 and 2005 thanks to high growth performance of countries like
Turkey, Uruguay, and Venezuela.
A simple measure of (apparent) productivity, real GDP per worker (ratio of GDP to labor
force), is calculated to understand possible differences between the two groups of
countries. Figure 2 compares the average productivity in the two groups between 1980
and 2005. While the productivity level of the second group of developing countries has
been almost flat, the fast-growing countries exhibit increasing productivity levels and the
level of productivity for this group passes the one in the comparison group after 1990.
The gap between the two groups continues to grow wider throughout the decades.
Together with increasing productivity in the fast-growing countries, we see that the share
of industrial production has also increased over time. Figure 3 summarizes changes in the
value added activities in the industrial sector for these two groups of countries. While the
share of industrial production is increasing and stays stable right above 40 percent of
GDP for the first group, this share drops to around 30 percent of GDP in the second
group of countries.
The second column of Tables 1 and 2 gives information about the size of public spending
as a share of GDP in the two groups of countries. While there are no sensible differences
in the size of the government budget between the two groups, an upward trend is
observed in the second group in recent years. For example it jumped to 33 percent in
Turkey on average between 2000 and 2005 and 31 percent in Uruguay.
Other than these trends, with the exception of Botswana, all countries in our sample have
managed to keep a relatively small size of total public spending, which is around below
30 percent of GDP,14 with the exception of Botswana at 38 percent. On the other side,
Singapore in the first sample, and the Philippines and Mexico in the second, managed to
keep the share of government expenditure in GDP relatively stable and at low levels
throughout the period analyzed. While Singapore has 18 percent public spending in
percent of GDP on average, Mexico and the Philippines have 17 and 16 percent
respectively.
Regarding the budget deficit, we observe that it is slightly larger for the comparison
group (-1.9 percent of GDP on average) compared to the one for the fast-growing
countries (-1.3 percent).
When we compare the share of productive expenditure in total public expenditure in
Tables 1 and 2, we can see that the share is significantly higher for the first group of
countries (62 percent), while it is only 50 percent for the second group throughout the
14
The definition used here refers to the consolidated central government only, which includes the central
government plus all government entities associated to it, and excludes all public spending at sub-national
level since it was not feasible to construct a reliable database for the consolidated general government
including all countries in the sample.
8
period included in the study.15 The other interesting result is that this share tends to
decline significantly for the second group (see Figure 4) especially after 1980. Thus, the
gap between the shares of productive expenditure in the two groups increases
significantly as time goes on, despite the fact that they were relatively close during the
mid to late 1970s.
Another major difference that can explain the gap between growth performances of the
two groups of countries may be in the shares of public gross fixed capital formation as
percent of GDP. Indeed, Figure 5 clearly shows that public investment is much larger in
the fast-growing countries. The share of public investment in GDP is around 10 percent
for the first group of countries, while it is only 5 percent for the second group. Given the
significance of public investment, especially its infrastructure component, in economic
growth, this difference between the two groups is striking. In addition, of course, there
may also be significant differences in the quality of capital spending, which may explain
differences in their impact on growth.
The differences in the impact of public spending on growth may also be associated with
the effectiveness and quality of governance. Figure 6 presents the percentiles for
government effectiveness.16 The figure shows that, in terms of government effectiveness,
all countries in the first group (with the exception of Indonesia) rank more favorably
when compared with the comparison group. When we check the second group (last seven
countries in Figure 6), we can see that the percentiles are much lower at around 60 with
the exception of Chile who has a government effectiveness index above 80.17 In the
whole sample of countries, Singapore enjoys the highest government effectiveness index.
On the other extreme case, Venezuela’s government effectiveness is in the bottom 16
percentile. Similarly, Figure 7 shows the differences in the bureaucracy quality between
the two groups. The gap between the two groups is obvious although it has gotten smaller
in recent years.
IV. EMPIRICAL SPECIFICATION AND ECONOMETRIC METHODS
Basic empirical specification
The empirical specification used in the analysis is similar to the one presented in MorenoDodson (2008). We add new techniques to control for possible differences between the
two groups of countries, in order to better understand the impact of public expenditure
and its components on growth. We also introduce nonlinearities in some specifications.
15
Productive expenditure is defined as the sum of general public services expenditure, defense expenditure,
educational expenditure, health expenditure, housing expenditure, transportation and communication
expenditure. See, for example, Bleaney, Kneller, and Glemmell (2001).
16
According to the KKM (Kraay, Kauffman and Mastruzzi) indicators, government effectiveness measures
the quality of public services, the quality of the civil service, the degree of its independence from political
pressures, the quality of policy formulation and implementation, and the credibility of the government’s
commitment to such policies.
17
Kray, Kauffman and Mastruzzi (KKM) indicators using a (0-100) percentile rank, World Bank. No data
exist before 1996.
9
The basic panel regression equation is first run separately for the fast-growing and the
comparison group18:
yˆ it  b1 yˆ it 1  b2 pit  b3 HC i  b4 FRit  b5 PEit  b6 FSit  b7 CPIINF
where:
i is the country index,
t is the year index,
ŷ is the rate of growth of GDP per capita,
p is the ratio of private investment to GDP,19
HC is the initial human capital index,
FR is the ratio of total fiscal revenues to GDP,
PE is the ratio of total public expenditures to GDP,20
FS is the ratio of the fiscal balance (deficit or surplus) to GDP,
CPIINF is the inflation rate,
and b1, b2, b3, b4, b5, b6, and b7 are the coefficients assigned to the independent variables.
The government budget constraint is included in the specification through revenues,
expenditures and fiscal balance. In some specifications total fiscal revenues are
disaggregated into tax and non-tax revenues.
18
Detailed explanation of the logic behind the specification is presented in Moreno-Dodson (2008).
In some regressions, openness has been used, instead of the private investment-to-GDP ratio, as a control
variable.
20
Unlike other studies testing only the impact of public investment on growth while ignoring current
spending, this analysis includes total public spending, capital and current, without specifically separating
them. The rationale for this decision is based on the evidence that some categories of current spending
items are indeed critical to ensure the profitability of investments. For example, operations and
maintenance expenditures, which are considered as current spending items, are critical to ensure the
profitability of infrastructure investments since they can facilitate access and prevent accidents, permitting
citizens to arrive safely to markets, schools, hospitals or any other destinations. Similarly, salaries of
teachers, usually classified under the current spending rubric, are closely connected to the quality of
education provided. In addition, it would not be realistic to try and isolate public investments completely
since in many countries capital budgets include de facto, explicitly or implicitly, salaries and current
spending items.
19
10
Total public spending is disaggregated using a definition based on Bleaney, Kneller, and
Glemmell (2001), and Kneller, Bleaney and Gemmell (1998), which classifies public
spending as productive versus unproductive components (a priori). As noted by these
authors, government spending items are classified according to whether they are to be
included in the private sector production function or not21. They are considered as
productive if they are considered to be used by the private sector in its production
function. If not, they are named as unproductive, a priori, meaning that they are not
expected to have an effect on the steady-state rate of growth:
Productive expenditure: General public services expenditure, Defense22
expenditure, Educational expenditure, Health expenditure, Housing expenditure,
Transportation and communication expenditure
Unproductive expenditure: Social security and welfare expenditure, Expenditure
on recreation, Expenditure on fuel and energy, Expenditure on agriculture,
forestry, fishing and hunting, Expenditure on mining, mineral resources,
manufacturing, and construction, Expenditure on other economic affairs and
services.
Empirical specification capturing differences in the two groups of countries
In an alternative empirical specification which is used when all countries in the data set
are pooled together, interactive dummy variables are introduced to capture possible
differences between fast-growing countries and the comparison group:
yˆit  b1 yˆit1  b2 pit  b3 HCi  b4 FRit  b5 DFAST * PEit  b6 DSLW * PEit  b7 FSit  b7CPIINF
where:
DFAST is the dummy variable which is 1 for fast-growing countries and 0 otherwise,
DCOMP is the dummy variable which is 1 for the other countries and 0 otherwise.
The multiplication of these dummy variables with public expenditure produces the
interaction variables that capture the possible impact of public expenditure on growth for
two separate groups.
Nonlinear empirical specification
In a dynamic setting and using recent econometric techniques, we also try to estimate
possible nonlinear effects of public expenditure on growth in fast-growing countries
versus the comparison group.23 The nonlinear empirical specification together with
interactive dummy variables capturing group effects is:
21
See Barro and Sala-i-Martin (2003).
See page 18 for a slightly different classification that excludes defense from productive spending.
23
As specified by Barro (1990), and as discussed earlier, the nonlinearity in the impact of public
expenditure on growth can be significant to understand the link between these two variables.
22
11
yˆ it  b1 yˆ it 1  b2 pit  b3 HC i  b4 FRit  b5 DFAST * PEit  b6 DFAST * PE 2 it
 b7 DCOMP * PEit  b8 DCOMP * PE 2 it  b9 FSit  b10CPIINF
where the square terms of total or productive component of government expenditure
capture the nonlinearity.24
Econometric methodology
Three alternative econometric methods for panel data are used (OLS, SURE, and GMM)
and their results are then compared.25 For each empirical specification, the first sets of
results are obtained using OLS and/or SURE methods. Then, a dynamic panel technique
(GMM) is applied and the results are compared with those obtained with the static panel
regressions. In the empirical analysis, due to data limitations, annual data are used instead
of longer-term averages, thus basically the focus is on the short-term growth impact of
government spending.26
OLS and SURE methods are based on the assumption that the right-hand-side variables
are exogenous. But it is quite likely that these variables may not be because they can be
determined by each other, or by the growth rate, or by other variables that are not
controlled for in the empirical specifications. The GMM dynamic panel method is used to
allow for a more rigorous treatment of the endogeneity of public spending with respect to
growth in order to have more reliable and precise results.27 More specifically, we use a
two-step GMM methodology, taking first differences of the variables. Since a set of
instrumental variables is used with the GMM technique, it helps us control for possible
endogenity among regressors.
In the regressions, the set of instruments consists of lagged values of dependent and
independent variables. The complete set of instrumental variables is:
 the second and third lags of the growth rate of GDP per capita, initial human
capital (or initial life expectancy or initial GDP per capita),
24
The dummy variables (DFAST and DCOMP, as defined above) control for the group effects. Of course,
there could also be nonlinearities related to other variables, most particularly private investment or
differences in inflation; it could be, for instance, that countries in the fast-growing group also have higher
rates of private investment and better macroeconomic policies. However, given the size of our sample, and
the issue at hand, we limit ourselves to studying nonlinearities associated with government spending.
25
See Moreno-Dodson (2008) for a detailed description of the three methods.
26 The SURE methodology used in the paper is a type of OLS technique which can be estimated to account
for various patterns of correlation among the residuals. In the paper, the variance structure introduced by
the SURE methodology is cross-section specific heteroskedasticity. This methodology is used to check
whether or not our results with ordinary least square change when we introduced cross-section specific
heteroskedasticity, given that the countries in the data set are different from each other.
27
The dynamic general method of moments (GMM) was introduced by Arellano and Bond (1991). As
indicated in the next section, OLS and SURE methodologies produce similar results, and GMM confirms
most of them.
12
 the first, second, and third lags of private investment in percent of GDP,
 the second and third lags of tax revenue, total public expenditure or productive
expenditure, unproductive expenditure, all in percent of GDP,
 the first, second, and third lags of other revenue and budget balance, all as a
share of GDP.
V. REGRESSION ANALYSIS
Each empirical specification introduced in the previous section is run in four different
settings, using the three econometric methods defined above. The first set of empirical
estimations, which is presented in the first column of the tables, is based on regressions
with the dataset for fast-growing countries.28 The second set of results (see the second
column of the tables) is for the comparison group. The third set of results, presented in
the third column of the tables, is obtained by combining the two groups of countries in a
panel data setting. Then the fourth set of results, as presented in the last column of the
tables, is obtained again by combining all countries in a single dataset but also by
introducing interactive dummy variables for public expenditure, to control for possible
group effects.
The overall results suggest that total public spending, especially its productive
component, has a statistically significant, positive impact on the GDP per capita growth
rate for fast-growing countries, while a similar link for the comparison group cannot be
established robustly.
V.1. Linear regression results
1.1 Total Public Spending
The linear regression specifications involve public spending and its components as
percents of GDP, ignoring any higher order variables. First we include total public
spending in the regressions. One common finding is that the differences in the size of
estimated coefficients for the fast-growing group versus the comparison group, as well as
their level of statistical significance, are both large.
Table 3 presents the first set of results obtained with both OLS and SURE methodologies.
For the fast-growing countries in our dataset (first two columns in Table 3), all the
estimated coefficients of government budget components including public spending are
statistically and economically significant determinants of economic growth at 1 percent
significance level. Both OLS and SURE results produce the same results. However, we
do not see similar results for the comparison group (columns 3 and 4 in Table 3). Only
the budget surplus has the expected positive sign and statistically significant at 5 percent
level. The impact of public spending on GDP per capita growth is positive but the
28
It should be noted that these first sets of results are the ones also presented in Moreno-Dodson (2008)
since we use the same fast growing countries in our analyses.
13
coefficient is not significant using neither OLS nor SURE methodologies. Similarly, total
fiscal revenue as a share of GDP does not have any significant impact on growth.29
When we compare the magnitudes of the coefficients, they are much larger for the first
group, indicating the economic importance of budget components on economic
development. For example, a one-percent increase in the public spending to GDP ratio
leads to almost half a percent increase in GDP per capita growth rates for the fastgrowing group, while the same increase in spending causes only a 0.04 percent increase
in GDP per capita growth in the comparison group.
Although the focus of this paper is on the link between public spending and growth, it is
also worth acknowledging the contribution of private investment. The presence and
significance of this variable is critical to assess whether or not complementary effects
between public expenditure and private investment have been triggered, leading to higher
growth. As indicated in Table 3, the empirical results state that private investment is
significant only for the fast-growing group. Similarly, inflation, which is included to
capture the impact of macroeconomic stability on growth, is a statistically significant
determinant of growth (with a negative coefficient) only for the fast-growing group.
The other interesting result is that when we combine the two sets together, the economic
and statistical significance of total public expenditure in determining growth drops
substantially. Columns 5 and 6 of Table 3 show the results when two country groups are
combined in one data set. In this case, the coefficient of total public spending is
significant only at 10 percent and only with the SURE methodology. The magnitude of
the coefficients is much smaller compared to the one produced when we consider only
fast-growing countries (columns 1 and 2 in Table 3). Similarly, other components of the
budget – fiscal revenue and budget surplus – are significant at 10 and 5 percent,
successively.30 Private investment continues to have a statistically significant coefficient
at the 1 percent level, indicating that the fast-growing group dominates.
These results lead us to conclude that when more heterogeneous countries (in terms of
their growth performance) are all included in the dataset, the significance of public
spending and other budget components drops. This may partially explain why some
empirical studies in the literature mixing countries with very different growth patterns
cannot find a statistically significant link between government spending and economic
growth. In other words, in order to avoid parameter bias in cross-country regression
results, distinguishing “clusters” by either specific characteristics and performance, or
more formal pooling tests, can be critical.
As explained in the empirical specification section of the paper, interactive dummy
variables are introduced to capture the possible differences in the impact between the two
groups in the combined panel regressions. The interactive dummy variable for the fast29
The lack of significance or robustness of the initial human capital variable may be due to the fact that the
indicator that we use does not capture well quality differences across countries and over time; see Rogers
(2008) for a more detailed discussion.
30
This finding supports the results presented in Lee and Gordon (2005).
14
growing countries (DFAST × public spending) has an economically and statistically
significant coefficient at 5 percent with OLS, while the similar interactive dummy
variable for the comparison group leads to a lower estimated coefficient and less
statistical significance (see columns 7 and 8 in Table 3). The results in this setting clearly
suggest that the impact of total public expenditures on growth is much more significant
for fast-growing countries when we use the combined dataset, after controlling for other
macroeconomic variables and budget components.
The estimated coefficients using OLS and SURE methodology are confirmed also by a
dynamic GMM methodology as presented in Table 4. On the one hand, total fiscal
spending is again a statistically and economically significant determinant of growth for
the fast-growing group (see column 1). Inflation also has expected signs and statistically
significant coefficients at 1 percent level.
Openness, which is included here to capture possible effects of private sector transactions
on growth through exports and imports (instead of private investment), and it is expected
to have a positive impact on growth, has indeed a statistically significant coefficient at 1
percent level.31
On the other hand, none of the variables are statistically significant in the case of the
comparison group, except inflation (see column 2). Total public spending even has an
unexpected negative impact on growth but it is not statistically significant. The
insignificant coefficient of openness for the comparison group again shows that export
and import flows are not successful in explaining growth for this group.
When the two sets are combined, the significance of total public spending in determining
growth disappears and the sign of the variable turns negative. But when the group effects
in the combined data setting are controlled for with the interactive dummies used with
total fiscal spending, it can be seen that the fast-growing group has a statistically
significant public spending impact on growth, while for the comparison group, the
influence of public spending on growth is negative (see column 4).
1.2. Expenditure Classifications: Productive versus Unproductive
The remaining question is how the impact of different components of public spending on
growth varies in these two groups of countries. The classification of public spending
considers productive versus unproductive public spending, as explained in the previous
section. 32
31
In each GMM specification, we include openness instead of private investment since this variable
produces better results.
32
According to the “a priori” definition introduced by Bleaney, Kneller, and Glemmell (2001) as explained
before, it is expected that the productive component of public spending will have a higher impact on
growth relative to the unproductive component. When we focus on fast-growing countries, the result is as
expected (column 1 in Table 5). In an empirical specification where we control for some macroeconomic
variables and lagged value of growth, fiscal revenue and budget balance, productive public spending has a
15
Column 3 of Table 5 shows that productive expenditure gets insignificant in explaining
growth in the pooled sample, again illustrating the caveats of pooling a disparate group of
countries. The estimation result with the interactive dummies for the two groups as
presented in column 4 of Table 5 also supports the previous findings: the coefficient of
the interactive dummy variable for productive expenditure in the fast-growing group is
statistically significant at 1 percent, while the one in the comparison group is not
significant.
The estimation results with the composition of public spending, based on the dynamic
GMM technique, are given in Table 6. Since they are similar to previous results obtained
with the panel OLS methodology, this suggests that the findings are robust across
different econometric methods. Again for fast-growing countries, productive public
spending has a positive and statistically significant impact at 10 percent on growth, which
is lower than the significance level reported in Table 5. The sign of the non-productive
component for the fast-growing group becomes negative in the GMM specifications, but
it does not have any statistically significant influence on growth even though it was
positively significant at the 10 percent level in Table 5 with the OLS specification.
Openness continues to have a statistically significant impact on growth at the 10 percent
level.
The comparison group, on the other hand, does not show any significant coefficient for
any component of government spending. The nonproductive component of government
spending continues to produce statistically insignificant coefficients. When we combine
two groups in one set, the sign of productive public expenditure is positive and its
significance is close to 10 percent. The nonproductive component turns out to have a
statistically significant, negative sign in the GMM specification. When we separate the
effects of two groups in the combined dataset with interactive dummy variables (Column
4 in Table 6), it can be seen that the positive impact of productive public spending on
growth in the combined set is confirmed only for the fast-growing country group, as was
the case with OLS.
V.2. Nonlinear regression results
As argued by Barro (1990, 1991), the link between public spending and growth is
expected to be positive when the size of government is small but it may turn negative as
the size gets larger. To capture nonlinearity, the additional explanatory variable
introduced in the nonlinear specification is either the squared value of total public
spending or the squared term of productive public spending, depending on which
specification is used. The OLS panel regression methodology is first used to test
nonlinearity. Then we also use the dynamic GMM methodology to confirm the results.
higher impact on growth in the fast-growing group, while it has a negative but insignificant impact for the
comparison group (compare columns 1 and 2 in Table 5).
16
Table 7 presents the coefficients estimated by the panel OLS methodology. In the table,
total public spending is considered. In none of the specifications, neither total public
expenditure nor its square term has a statistically significant effect on growth. This result
is consistent across the country groups and the combined set. When we control for the
group effects with the help of interactive dummy variables in the combined set, not much
difference is observed. These findings do not change with any alternative econometric
technique. Table 8 shows the same specifications estimated this time with the GMM
technique. The results are robust: total public expenditure and its square term again are
not significant in any column of Table 8.
As presented previously, the empirical results in the linear setting suggest that some
components of public spending are more influential in determining the growth rate of
GDP per capita. Thus, we expect that the inclusion of those components of public
spending instead of total public spending may matter in the nonlinear setting as well.
Table 9 shows the estimated coefficients with the level and the squared value of
productive public expenditure, using panel OLS estimation technique. As was the case
before, the productive component of government spending is a statistically significant
determinant of the growth rate of GDP per capita only for fast-growing countries. The
squared value of productive expenditures has a negative sign, but it is not statistically
significant. Table 10, which uses the GMM estimation technique, confirms the results
presented in Table 9. Similarly, the productive component is significant only for fastgrowing countries, while the squared value of this variable is not statistically significant
in any case.
Overall the results based on the nonlinear specifications do not identify any nonlinearity
between total public spending or its productive components and economic growth.
V.3. Robustness check of the classification of public spending
The inclusion of defense spending item in the a priori definition of productive public
spending can raise some questions. There is still a controversy on whether defense
spending promotes economic growth. For example, Benoit (1973 and 1978) explains that
defense expenditures can increase growth through contributing to providing education
and health services to staff in the military, lowering unemployment, engaging in public
works, and increasing scientific and technological innovations. It may also promote
growth by providing a more secure environment for private investors. However, another
group of papers argues that the relationship between defense spending and growth is not
robust across different countries.33
Since we do not have much knowledge of the defense sector and to test the robustness of
the empirical results to alternative classifications, we rerun our regressions by excluding
defense spending from productive public spending. It is added to “other” public spending
and therefore excluded from the regression. Table 11 shows the empirical results with
33
For example, see Biswas and Ram (1986), and Looney and Frederiksen (1986).
17
this alternative definition of a priori productive spending.34 When we compare both
findings, we can see that the original results are robust. The productive spending group is
still a statistically and economically significant determinant of growth only for the fastgrowing group. It is not significant for the comparison group, or when we pool all
countries together.
VI. CONCLUSIONS
The main purpose of this paper has been to empirically investigate the link between
public spending and its components, and growth with a panel dataset where developing
countries are classified according to their GDP per capita rates of growth in two groups:
fast-growing countries and a comparison group including a mix of countries with
different growth patterns (a lower growth rate on average).
The main result of the study is that the influence of public spending on economic growth
is clearly different in the two groups, even though the size of government measured as
total public expenditures as percent of GDP is similar. The main explanation is found in
the composition of public expenditure.35 On average, the fast-growing group has a much
higher share of productive expenditure in total, while unproductive components of public
spending are larger for the comparison group.
In the fast-growing countries, the productive components of public spending consistently
have a positive and statistically significant impact on economic growth, after controlling
for macroeconomic and private sector variables, initial conditions of countries, and other
budget components. This statistically significant effect cannot be established in a robust
fashion for the comparison group. Our results are consistent with several other studies
focusing on developed and developing countries, such as Bose, Haque, and Osborn
(2007), and Benos (2009).36
In addition to the differences that we observe in the response of growth to the
composition of public spending between two groups of countries, we also find
differences in the impact of private sector and macroeconomic stability. Two alternative
indicators of private sector involvement used in the specifications, private investment and
openness, tend to have higher significance in explaining growth for the fast-growing
group. One possible explanation can be that the expected complementarity effect between
public and private sector in the growth process is only manifested in the first group.
Another reason can be differences in investment climate between the two groups.
34
Due to space limitations, they are not included in the paper, but other empirical specifications are also
robust to the alternative definition of productive public spending.
35
We also observe some differences in government effectiveness and bureaucracy quality. Due to data
limitations, we could not explicitly include these variables in our regressions. But because the limited data
that we could find indicate that the quality of governance (as measured by government effectiveness and
bureaucracy quality) is consistently higher for the fast-growing group, we believe that the group effects that
are introduced in the empirical specification partially capture the quality of governance.
36
For dissenting results, however, see Ghosh and Gregoriou (2008).
18
Similarly, inflation is negatively correlated with growth only for the fast-growing group,
indicating that reducing inflation leads to faster growth and therefore growth is
responsive to improvements in macroeconomic stability in those countries.
When we combine the two groups of countries in the regression analysis, the link
between government spending and growth again gets weaker or disappears. In the
regressions with combined data, when we control the group effects by interactive dummy
variables, it can be seen that only the fast-growing panel produces a positive and
statistically significant link between the productive component of public spending and
growth. Our overall conclusion therefore is that public spending can be a significant
determinant of growth, if countries are able to devote a significant fraction of these
expenditures to productive uses.
19
REFERENCES
Agell, J., H. Ohlsson, and Skogmann Thoursie P., 2006, “Growth Effects of Government
Expenditure and Taxation in Rich Countries: A comment,” European Economic Review,
50, 211-18.
Agénor, Pierre-Richard, The Economics of Adjustment and Growth, Harvard University
Press (Cambridge, Mass.: 2004).
------, "Fiscal Policy and Endogenous Growth with Public Infrastructure," Oxford
Economic Papers, 60 (January 2008), 57-88.
Agénor, Pierre-Richard, and Blanca Moreno-Dodson, "Public Infrastructure and Growth:
New Channels and Policy Implications," in Public Expenditure, ed. By Maura Francese,
Daniele Franco, and Raffaela Giordano, Banca d'Italia (Rome: 2007); and WB Working
Paper.
Agénor, Pierre-Richard, and Kyriakos Neanidis, "The Allocation of Public Expenditure
and Economic Growth," Working Paper No. 69, Centre for Growth and Business Cycle
Research, University of Manchester (March 2006).
Ang, James B., 2009, “Do public investment and FDI crowd in or crowd out private
domestic investment in Malaysia?” Applied Economics, 41, 913–919.
Arellano, M. and S. Bond, 1991, “Some Tests of Specification for Panel Data:
Montecarlo Evidence and Application to Employment Equations,” Review of Economic
Studies 58(2), 277-297.
Barro, Robert J., "Government Spending in a Simple Model of Endogenous Growth,"
Journal of Political Economy, 98 (October 1990), s103-s25.
Barro, Robert J., "Economic Growth in a Cross Section of Countries,” Quarterly Journal
of Economics, vol. 106 (May 1991), 407-43.
Barro, Robert J. and Xavier Sala-i-Martin, 1992, “Public Finance in Models of Economic
Growth,” Review of Economic Studies, vol. 99, No.4, 645-61.
Barro, Robert J., and Xavier Sala-i-Martin, Economic Growth, 2nd. ed., McGraw-Hill
(New York: 2003).
Benoit, E., 1973, Defense and Economic Growth in Developing Countries. Lexington:
Lexington Books.
Benoit, E., 1978, “Growth and defense in developing countries.” Economic Development
and Cultural Change 26(January): 271-280.
20
Benos, Nikos, 2009, “Fiscal Policy and Economic Growth: Empirical Evidence from EU
Countries,” unpublished, University of Ioannina (September 2009).
Biswas, B. and R. Ram (1986) “Military expenditures and economic growth in less
developed countries: an augmented model and further evidence.” Economic Development
and Cultural Change 34(January): 361-372.
Bleaney, Michael, Norman Gemmell, and Richard Kneller, 2001, “Testing the
Endogenous Growth Model: Public Expenditure, Taxation, and Growth over the Long
Run,” Canadian Journal of Economics, Vol. 34, pp.36-57.
Bose, N., Haque M.E., and Osborn, D. R., 2007, “Public Expenditure and Economic
Growth: A disaggregated analysis for developing countries,” The Manchester School, 75,
533-56.
Colombier, C., 2009, “Growth effects of fiscal policies: an application of robust modified
M-estimator,” Applied Economics, 41, 899–912.
Easterly, William and Sergio Rebelo, 1993, “Fiscal Policy and Economic Growth: An
Emperical Investment,” Journal of Monetary Economics, vol. 32, no: 3, pp. 417-58.
Folster, Stefan and Magnus Henrekson, 2001, “Growth Effects of Government
Expenditure and Taxation in Rich Countries,” European Economic Review, vol. 45(8),
1501-20.
Ghosh, Sugata, and Andros Gregoriou, 2008, “The Composition of Government
Spending and Growth: Is Current or Capital Spending Better?,” Oxford Economic
Papers, 60 (June), 484-516.
Glomm, Gerhard and B. Ravikumar, 1997, “Productive Government Expendituers and
Long-run Growth,” Journal of Economic Dynamics and Control, vol. 21, No.1, 183-204.
Grossman, Philip J., 1990, “Government and Growth: Cross-sectional Evidenece,” Public
Choice Vol.65: 217-227.
Grossman, Philip J., 1988, “Government and economic growth: A nonlinear
relationship,” Public Choice vol. 56(2): 193-200.
Gupta, Sanjeev, Benedict Clements, Emanuele Baldacci, and Carlos Mulas-Granados,
2002, “Expenditure Composition, Fiscal Adjustment, and Growth in Low-income
Countries,” IMF Working paper 02/77 (Washington DC).
Kellenberg, Derek K., 2009, “US Affiliates, Infrastructure and Growth: A Simultaneous
Investigation of Critical Mass,” Journal of International Trade and Economic
Development, 18 (September), 311-45.
21
Kneller, Richard, Michael F. Bleaney, and Norman Gemmell, 1998, “Growth, Public
Policy and the Government Budget Constraint: Evidence from OECD Countries,”
Discussion Paper No. 98/14, School of Economics, University of Nottingham.
Kneller, Richard, Michael Bleaney, and Norman Gemmell, 1999, “Fiscal Policy and
Growth: Evidence from OECD Countries,” Journal of Public Economics, vol. 74, no. 2,
171-190.
Lee, Young and Roger H. Gordon, 2005, “Tax Structure and Economic Growth,” Journal
of Public Economics, 89:1027-1043.
Levine, Ross and David Renelt, 1992, “A Sensitivity Analysis of Cross-Country Growth
Regressions,” American Economic Review, vol. 82, no. 4, 942-63.
Looney, Robert and P.C. Frederiksen, 1986, “Profiles of Current Latin American Military
Producers”, International Organization, 40(3): 745-752.
Moreno-Dodson, Blanca, 2008, “Assessing the Impact of Public Spending on Growth:
An Empirical Analysis for seven fast Growing Countries,” World Bank Policy Research
Working Paper No. 4663.
Park, Hyun, 2006, “Expenditure Composition and Distortionary Tax for Equitable
Economic Growth,” IMF Working Paper No: WP/06/165.
Pushak, Tearas, Erwin R. Tiongson, and Aristomene Varoudakis, 2007, “Public Finnace,
Governance, and Growth in Transition Economies,” World Bank Policy Research
Working Paper No. 4255.
Rogers, Mark L., 2008, “Directly Unproductive Schooling: How Country Characteristics
Affect the Impact of Schooling on Growth,” European Economic Review, 53, 356-85.
Schaltegger, Christoph A. and Benno Torgler, 2006, “Growth effects of public
expenditure on the state and local level: evidence from a sample of rich governments,”
Applied Economics, 2006, 38, 1181–1192.
Slemrod, J., 1995, “What Do Cross-Country Studies Teach about Government
Involvement, Prosperity, and Economic Growth?” Brookings Papers on economic
Activity, pp.373-431.
Wahab, M., 2004, “Economic growth and Government Expenditure: Evidence from a
new test specification,” Applied Economics, 36, 2125-35.
Zagler, Martin, and Georg Durnecker, "Fiscal Policy and Economic Growth," Journal of
Economic Surveys, 17 (July 2003), 397-418.
22
APPENDIX – INFORMATION ON DATA
COUNTRIES INCLUDED
Fast growing countries: Botswana, Indonesia, Korea, Malaysia, Mauritius, Singapore,
Thailand.
Comparison group: Chile, Costa Rica, Mexico, Philippines, Turkey, Uruguay and
Venezuela
YEARS INCLUDED BY COUNTRIES
The years included are according to the data availability of government budget items.
Botswana: 1970-96 and 2001-2005
Indonesia: 1972-1999 and 2001-2005
Korea: 1970-97 and 2001-2005
Malaysia: 1972-87 and 2001-2005
Mauritius: 1973-2005
Singapore: 1972-2005
Thailand: 1972-2001
Chile: 1972-2001
Costa Rica: 1972-2001
Mexico: 1972-2000
Philippines: 1972-2001
Turkey: 1970-2001
Uruguay: 1972-2001
Venezuela: 1970-2001
LIST OF VARIABLES
1. Control Variables
Private Investment: Gross fixed capital formation by the private sector includes land
improvements (fences, ditches, drains, and so on); plant, machinery, and equipment
purchases; and the construction of roads, railways, and the like, including schools,
offices, hospitals, private residential dwellings, and commercial and industrial buildings.
Source: WDI and IFC working paper.
Imports of goods and services (% of GDP): Imports of goods and services represent the
value of all goods and other market services received from the rest of the world. They
include the value of merchandise, freight, insurance, transport, travel, royalties, license
fees, and other services, such as communication, construction, financial, information,
business, personal, and government services. They exclude labor and property income
(formerly called factor services) as well as transfer payments.
Source: WDI.
23
Exports of goods and services (% of GDP): Exports of goods and services represent the
value of all goods and other market services provided to the rest of the world. They
include the value of merchandise, freight, insurance, transport, travel, royalties, license
fees, and other services, such as communication, construction, financial, information,
business, personal, and government services. They exclude labor and property income
(formerly called factor services) as well as transfer payments.
Source: WDI.
Openness (% of GDP): Exports of goods and services (% of GDP) plus imports of
goods and services (% of GDP).
Bureaucracy Quality Index: Indexed series between 1 and 4 measuring the quality of
bureaucracy. Higher numbers indicate higher quality.
Source: International Country Risk Guide.
2. Initial Condition Variables
GDP per capita in constant 2000 U.S. dollar: GDP per capita is gross domestic product
divided by midyear population. GDP is the sum of gross value added by all resident
producers in the economy plus any product taxes and minus any subsidies not included in
the value of the products. It is calculated without making deductions for depreciation of
fabricated assets or for depletion and degradation of natural resources. Data are in
constant U.S. dollars.
Source: WDI.
Initial Human Index: Following Bose, Haque, and Osborn (2007), we construct the
initial human capital variable as the weighted sum of the initial enrolment ratios (%) in
primary and secondary schools and in higher education. The weights are 1 for primary
school enrolment ratio, 2 for secondary school and 3 for enrolment in higher education.
The weights are approximations to the relative values of three types of education. The
initial year is 1970.
Source: Barro and Sala-i-Martin (1995, 1999).
3. Government Budget Items
Tax Revenue: Tax revenue refers to compulsory transfers to the central government for
public purposes. Certain compulsory transfers such as fines, penalties, and most social
security contributions are excluded. Refunds and corrections of erroneously collected tax
revenue are treated as negative revenue. For central government.
Source: GFS.
Total Government Revenue: Revenue is cash receipts from taxes, social contributions,
and other revenues such as fines, fees, rent, and income from property or sales. Grants are
also excluded here. For central government.
Source: GFS.
24
Other Revenue: Total government revenue minus tax revenues. For central government.
Source: Calculated by author.
Total Government expenditure: Total central government expenditure: economic plus
social, plus general public services, plus defense, plus other expenditure.
Source: GFS.
Budget balance: Before 2001, it is calculated as total government revenue plus grants
minus total expenditure minus net lending. After 2001, it is total revenue plus grants
minus total expenditure. For central government.
Source: Calculated by author.
Productive expenditure:
General public services expenditure
Defense expenditure
Educational expenditure
Health expenditure
Housing expenditure
Transportation and communication expenditure
Source: Calculated by author.
Unproductive expenditure:
Social security and welfare expenditure
Expenditure on recreation
Expenditure on fuel and energy
Expenditure on agriculture, forestry, fishing and hunting
Expenditure on mining, mineral resources, manufacturing, and construction
Expenditure on other economic affairs and services
Source: Calculated by author.
Other expenditure:
Public order and safety
Other expenditures
Source: Calculated by author.
25
Correlation matrix
In order to give some initial idea about the possible links between public spending and its
components and economic growth, a simple correlation coefficient matrix is presented in
Table A.1. The correlation may change in the dynamic settings of the econometric
analysis, but overall we can see that the simple relationship between GDP per capita
growth and public expenditure (and its components) is much weaker and even negative in
some cases for the second group of countries. For the first group, the highest correlation
coefficients for growth belong to productive expenditures, economic expenditures, and
transportation expenditures. All these correlation coefficients are negative but close to
zero for the second group, indicating almost no correlation between the components of
public spending and growth.
Table A1 - Correlation matrix of variables
SET OF FAST GROWING COUNTRIES
Pub
spending
GDPPCGR /GDP
GDPPCGR
Pub spending/GDP
Fiscal balance/GDP
Pro/TEXP
Edu/TEXP
Hea/TEXP
TRA/TEXP
1.000
0.126
0.088
0.333
0.005
-0.043
0.383
1.000
-0.067
-0.038
0.231
0.396
0.103
SET OF OTHER DEVELOPING COUNTRIES
Pub
spending
GDPPCGR /GDP
GDPPCGR
Pub spending/GDP
Fiscal balance/GDP
Pro/TEXP
Edu/TEXP
Hea/TEXP
TRA/TEXP
1.000
-0.035
-0.205
-0.058
0.007
-0.027
-0.003
1.000
-0.401
-0.539
-0.563
0.010
-0.574
Fiscal
balance/GDP Pro/TEXP
1.000
0.487
0.273
-0.144
0.044
1.000
0.593
0.184
0.456
Fiscal
balance/GDP Pro/TEXP
1.000
0.389
0.289
0.283
0.264
1.000
0.484
0.276
0.680
Edu/TEXP
1.000
0.491
0.140
Edu/TEXP
1.000
0.313
0.254
Hea/TEXP
1.000
-0.053
Hea/TEXP
1.000
-0.041
TRA/TEXP
1.000
TRA/TEXP
1.000
Note: GDPPCGR is the growth rate of GDP per capita in real terms. Pub spending/GDP is the ratio of total public
spending in %. Fiscal balance is the ratio of fiscal balance to GDP in %. Pro/TEXP is public productive expenditure in
total public expenditure in %. Edu/TEXP is public education expenditure in total public expenditure in %. Hea/TEXP is
public health expenditure in total public expenditure in %. Tra/TEXP is public transportation expenditure in total
public expenditure in %.
26
TABLES
Table 1 - SET OF FAST GROWING COUNTRIES
Pub
Fiscal
spending balance/
GDPPCGR /GDP
GDP
Pro/TEXP Edu/TEXP Hea/TEXP TRA/TEXP
Botswana
Botswana
Botswana
Botswana
Indonesia
Indonesia
Indonesia
Indonesia
Korea, Rep.
Korea, Rep.
Korea, Rep.
Korea, Rep.
Malaysia
Malaysia
Malaysia
Malaysia
Mauritius
Mauritius
Mauritius
Mauritius
Singapore
Singapore
Singapore
Singapore
Thailand
Thailand
Thailand
Thailand
AVERAGE
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
11.93
8.00
3.74
5.60
5.31
4.41
3.26
3.37
6.32
6.38
5.24
4.60
5.18
3.18
4.52
3.15
4.90
4.24
3.15
7.55
5.32
4.39
3.44
4.83
5.48
3.99
4.08
5.02
28.37
32.53
36.91
38.83
18.36
21.17
17.23
18.33
15.48
15.91
16.39
20.71
23.70
30.58
24.23
26.99
30.16
26.81
23.59
24.36
18.07
24.99
15.60
16.58
15.49
18.39
17.61
19.59
22.75
-5.03
9.65
7.60
0.53
-3.35
-3.59
-0.85
-1.31
-0.79
0.19
1.21
1.83
-6.33
-9.55
-6.48
-11.28
-6.62
-3.26
-3.32
0.87
2.68
7.80
4.73
-2.71
-3.53
-0.25
-3.41
-1.28
76.61
68.87
74.11
62.79
63.20
53.02
15.72
68.33
65.52
55.99
59.87
58.93
57.90
60.92
61.09
51.42
46.29
45.85
52.89
79.37
71.59
90.26
91.49
62.21
57.40
61.35
59.96
61.96
16.54
19.23
23.01
23.52
8.42
9.15
8.72
4.88
16.21
18.66
18.90
15.38
22.45
18.80
20.95
24.60
14.10
14.86
15.95
15.74
16.74
18.62
24.24
23.70
20.70
19.64
20.38
21.30
17.69
5.66
5.52
5.22
8.48
2.03
2.18
2.54
1.33
1.29
1.65
1.04
0.45
6.71
4.85
5.85
7.06
8.50
7.78
8.56
8.70
7.89
5.72
7.67
6.65
4.11
5.41
7.51
8.97
5.33
13.71
9.92
6.76
12.24
9.83
7.35
1.09
5.47
3.77
5.94
6.67
5.87
8.36
8.02
8.03
4.30
4.50
4.37
3.40
5.31
8.01
6.02
7.82
10.29
7.02
11.28
7.11
7.13
Note: GDPPCGR is the growth rate of GDP per capita in real terms. Pub spending/GDP is the ratio
of total public spending in %. Fiscal balance is the ratio of fiscal balance to GDP in %. Pro/TEXP is
public productive expenditure in total public expenditure in %. Edu/TEXP is public education
expenditure in total public expenditure in %. Hea/TEXP is public health expenditure in total public
expenditure in %. Tra/TEXP is public transportation expenditure in total public expenditure in %.
27
Table 2 - COMPARISON GROUP
Pub
Fiscal
spending balance/
GDPPCGR /GDP
GDP
Pro/TEXP Edu/TEXP Hea/TEXP TRA/TEXP
Chile
Chile
Chile
Chile
Costa Rica
Costa Rica
Costa Rica
Costa Rica
Mexico
Mexico
Mexico
Mexico
Philippines
Philippines
Philippines
Philippines
Turkey
Turkey
Turkey
Turkey
Uruguay
Uruguay
Uruguay
Uruguay
Venezuela
Venezuela
Venezuela
Venezuela
AVERAGE
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
1970-79
1980-89
1990-99
2000-05
0.86
2.72
4.66
3.09
3.70
-0.46
2.90
1.69
3.33
0.12
1.66
1.49
2.90
-0.45
0.53
2.57
2.28
1.70
2.03
3.60
2.33
0.07
2.60
0.71
0.45
-2.89
0.30
1.35
1.64
33.20
28.29
19.79
14.73
20.18
20.01
21.37
22.91
14.53
23.21
15.44
15.93
13.86
13.70
18.98
19.35
18.64
17.44
18.35
33.14
23.18
25.23
28.89
31.41
19.44
21.93
20.21
24.07
21.34
-1.85
-0.51
1.21
0.02
-2.75
-2.02
-1.62
-1.26
-2.48
-7.85
-0.69
-1.17
0.20
-0.42
-0.91
-3.92
-2.02
-3.35
-4.59
-11.18
-1.73
-2.46
-0.93
-4.05
4.26
1.94
0.11
-2.62
-1.88
58.77
50.04
67.87
70.40
54.62
53.15
41.47
31.65
43.54
48.53
70.39
73.91
50.69
38.94
66.64
71.98
59.12
26.38
45.95
38.62
28.51
26.06
54.77
49.30
41.52
49.34
50.47
14.24
13.41
14.62
18.02
28.00
20.14
19.18
20.98
17.92
12.97
23.37
24.73
14.53
17.23
17.57
17.95
19.50
13.10
14.49
8.78
10.64
7.34
6.91
7.33
16.47
18.55
17.03
21.22
16.29
7.41
7.23
11.47
12.54
9.32
25.26
23.20
22.03
4.28
1.58
3.25
4.95
4.42
5.01
3.17
2.10
2.98
2.25
3.28
3.24
4.19
4.14
5.49
6.17
9.99
8.62
7.05
6.65
7.55
4.75
3.29
14.51
9.67
4.84
6.10
9.50
5.03
4.23
2.32
17.15
21.18
11.08
8.54
15.46
8.20
5.11
2.32
5.20
6.29
4.45
2.89
9.75
6.19
2.38
2.69
7.43
Note: GDPPCGR is the growth rate of GDP per capita in real terms. Pub spending/GDP is the
ratio of total public spending in %. Fiscal balance is the ratio of fiscal balance to GDP in %.
Pro/TEXP is public productive expenditure in total public expenditure in %. Edu/TEXP is public
education expenditure in total public expenditure in %. Hea/TEXP is public health expenditure
in total public expenditure in %. Tra/TEXP is public transportation expenditure in total public
expenditure in %.
28
29
-0.205
(-5.65)***
0.406
(3.928)***
120
0.443
Inflation - consumer price index
Growth rate of GDP per capita (-1)
No. of observations
Adjusted R2
120
0.473
0.378
(4.11)***
-0.177
(-4.908)***
0.304
(3.942)***
…
-0.296
(-4.175)***
167
0.095
0.284
(3.611)***
-0.009
(-0.706)
0.250
(2.149)**
…
…
0.043
(0.808)
-0.064
(-0.745)
-0.008
(-0.732)
0.061
(0.557)
167
0.072
0.240
(3.237)***
-0.018
(-1.63)
0.090
(0.816)
…
…
0.026
(0.448)
0.002
(0.028)
-0.004
(-0.476)
0.080
(0.817)
(4)
287
0.312
0.296
(4.666)***
-0.023
(-2.4)**
0.161
(2.317)**
…
…
0.068
(1.481)
-0.084
(-1.765)*
-0.006
(-1.228)
0.143
(3.148)***
(5)
OLS
287
0.284
0.299
(5.141)***
-0.022
(-2.47)**
0.071
(1.365)
…
…
0.069
(1.896)*
-0.059
(-1.503)
-0.010
(-2.183)**
0.102
(2.652)***
(6)
SURE
Whole sample pooled
287
0.321
0.290
(4.609)***
-0.017
(-1.645)*
0.202
(2.825)***
0.059
(1.3)
0.159
(2.562)**
-0.129
(-2.492)**
0.001
(0.184)
0.087
(1.678)*
(7)
OLS
287
0.301
0.294
(5.116)***
-0.023
(-2.489)**
0.134
(2.545)**
0.063
(1.838)*
0.163
(3.298)***
-0.119
(-2.807)***
-0.002
(-0.399)
0.037
(0.86)
(8)
SURE
Whole sample with dummies
Note: The estimation method is the ordinary least squares f or panel data in the f irst column and the cross-section seeminly unrelated regression f or panel data in the
second column. Annual data are used. (-1) indicates variables lagged one period. DFAST is the dummy variable which is 1 f or f ast growing countries and 0 otherwise.
DCOMP is the dummy variable which is 1 f or the comparison group and 0 otherwise. t-statistics are given in parenthesis. * indicates 10% signif icance level, ** indicates
5% signif icance level, and *** indicates 1% signif icance level. These signif icance levels are equal to one minus the probabil ity of rejecting the null hypothesis of zero
coef f icients.
0.341
(3.668)***
Budget surplus
(% of GDP)
…
DFAST*(Productive plus unproductive expenditures (-1))
(% of GDP)
…
…
0.477
(3.876)***
Productive plus unproductive expenditures (-1)
(% of GDP)
DCOMP*(Productive plus unproductive expenditures (-1))
(% of GDP)
0.448
(4.263)***
-0.323
(-3.884)***
Total fiscal revenue
(% of GDP)
0.006
(0.931)
0.007
(0.96)
Initial human capital
0.115
(2.698)***
0.107
(1.917)*
(2)
(1)
(3)
OLS
SURE
OLS
SURE
Comparison group
Fast growing countries
Private investment
(% of GDP)
Dependent variable: Growth rate of GDP per capita
Table 3: Results with Total Public Expenditures (panel OLS and SURE)
Table 4: Results with Total Public Expenditures (Dynamic Panel - GMM)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
(1)
Comparison
group
(2)
Whole sample
pooled
(3)
Whole
sample with
dummies
(4)
Openness
(% of GDP)
0.062
(6.032)***
-0.006
(-0.073)
0.005
(0.141)
-0.004
(-0.123)
Total fiscal revenue
(% of GDP)
-0.207
(-0.662)
-0.136
(-0.146)
0.686
(0.94)
1.773
(1.523)
Total fiscal expenditures (-1)
(% of GDP)
0.318
(1.572)*
-0.326
(-1.091)
-0.019
(-0.117)
…
DFAST*(Total fiscal expenditures (-1))
(% of GDP)
…
…
…
1.698
(1.856)*
DCOMP*(Total fiscal expenditures (-1))
(% of GDP)
…
…
…
-0.918
(-1.72)*
Budget surplus
(% of GDP)
-0.020
(-0.097)
0.281
(0.275)
-0.164
(-0.321)
-0.946
(-1.479)
Inflation - consumer price index
-0.492
(-6.579)***
-0.056
(-4.782)***
-0.086
(-6.756)***
-0.088
(-3.671)***
Growth rate of GDP per capita (-1)
0.125
(0.906)
-0.092
(-1.675)*
-0.089
(-0.807)
-0.006
(-0.031)
No. of observations
J-statistics
94
7.542
169
2.461
264
2.001
264
1.301
Note: The estimation method is a dynamic GMM. Annual data are used. (-1) indicates variables lagged one
period. DFAST is the dummy variable which is 1 f or f ast growing countries and 0 otherwise. DCOMP is the dummy
variable which is 1 f or the comparison group and 0 otherwise. t-statistics are given in parenthesis. * indicates 10%
signif icance level, ** indicates 5% signif icance level, and *** indicates 1% signif icance level. These signif icance
levels are equal to one minus the probability of rejecting the null hypothesis of zero coef f icients. J -test is f or
overidentif ication problem where H0: there is no overidentif ication problem. We f ail to reject in each case.
30
Table 5: Results with Productive and Non-productive Expenditures (panel OLS)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
(1)
Comparison
group
(2)
Whole sample
pooled
(3)
Whole sample
with dummies
(4)
Private investment
(% of GDP)
0.071
(1.235)
0.111
(0.975)
0.154
(3.244)***
0.086
(1.629)
Initial human capital
0.007
(0.987)
-0.015
(-1.326)
-0.010
(-1.689)*
-0.005
(-0.747)
Tax revenue
(% of GDP)
-0.201
(-1.877)*
-0.038
(-0.313)
-0.023
(-0.287)
-0.037
(-0.438)
Other revenue
(% of GDP)
-0.531
(-4.051)***
-0.346
(-1.548)
-0.144
(-1.853)*
-0.242
(-3.143)***
Productive expenditure (-1)
(% of GDP)
0.664
(4.314)***
-0.051
(-0.574)
0.048
(0.698)
…
DFAST*Productive expenditure (-1)
(% of GDP)
…
…
…
0.219
(2.985)***
DCOMP*Productive expenditure (-1) …
(% of GDP)
…
…
0.153
(1.3)
Non-productive expenditure (-1)
(% of GDP)
0.312
(1.786)*
0.107
(0.944)
0.076
(0.904)
-0.060
(-0.657)
Budget surplus
(% of GDP)
0.450
(4.182)***
0.275
(2.318)**
0.184
(2.571)**
0.231
(3.106)***
Inflation - consumer price index
-0.208
(-5.76)***
-0.013
(-1.015)
-0.023
(-2.332)**
-0.019
(-1.813)*
Growth rate of GDP per capita (-1)
0.377
(3.644)***
0.255
(3.181)***
0.284
(4.442)***
0.264
(4.169)***
No. of observations
Adjusted R2
120
0.454
167
0.101
287
0.312
287
0.327
Note: The estimation method is the ordinary least squares f or panel data. Annual data are used. (-1) indicates
variables lagged one period. D-FAST is the dummy variable which is 1 f or f ast growing countries and 0 otherwise.
DCOMP is the dummy variable which is 1 f or the comparison group and 0 otherwise. t-statistics are given in
parenthesis. * indicates 10% signif icance level, ** indicates 5% signif icance level, and *** indicates 1% signif icance
level. These signif icance levels are equal to one minus the probability of rejecting the null hypothesis of zero
coef f icients.
31
Method: Pan
Date: 05/11/
Sample: 197
Table 6: Results with Productive and Non-productive Expenditures
(Dynamic Panel - GMM)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
(1)
Comparison
group
(2)
Whole sample
pooled
(3)
Whole sample
with dummies
(4)
Openness
(% of GDP)
0.058
(1.834)*
-0.348
(-1.335)
0.047
(0.936)
-0.014
(-0.17)
Tax revenue
(% of GDP)
-0.153
(-0.355)
1.481
(1.037)
0.154
(0.22)
0.377
(0.445)
Other revenue
(% of GDP)
-0.400
(-0.905)
-0.639
(-0.26)
-0.374
(-0.504)
-0.081
(-0.089)
Productive expenditure (-1)
(% of GDP)
0.777
(1.673)*
0.397
(0.393)
0.866
(1.513)
…
DFAST*Productive expenditure (-1)
(% of GDP)
…
…
…
3.217
(1.465)
DCOMP*Productive expenditure (-1)
(% of GDP)
…
…
…
-0.331
(-0.263)
Non-productive expenditure (-1)
(% of GDP)
-0.890
(-1.301)
-1.187
(-0.975)
-2.003
(-2.641)***
-2.181
(-2.402)**
Budget surplus
(% of GDP)
0.126
(0.316)
0.115
(0.13)
-0.058
(-0.108)
-0.017
(-0.027)
Inflation - consumer price index
-0.375
(-3.12)***
-0.105
(-2.743)***
-0.103
(-2.972)***
-0.116
(-2.735)***
Growth rate of GDP per capita (-1)
-0.03
(-0.151)
-0.378
(-1.436)
-0.382
(-1.749)*
-0.331
(-1.27)
No. of observations
J-test
98
1.993
171
2.07
269
6.97
269
3.75
Note: The estimation method is a dynamic GMM. Annual data are used. (-1) indicates variables lagged one
period. DFAST is the dummy variable which is 1 f or f ast growing countries and 0 otherwise. DCOMP is the
dummy variable which is 1 f or the comparison group and 0 otherwise. t-statistics are given in parenthesis. *
indicates 10% signif icance level, ** indicates 5% signif icance level, and *** indicates 1% signif icance level.
These signif icance levels are equal to one minus the probability of rejecting the null hypothesis of zero
coef f icients. J-test is f or overidentif ication problem where H0: there is no overidentif ication problem. We f ail to
reject in each case.
32
Table 7: Nonlinear Results with Total Public Expenditures (OLS)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
Comparison
group
Whole sample
pooled
Whole sample
with dummies
Private investment
(% of GDP)
0.091
(1.584)
0.076
(0.673)
0.151
(3.241)***
0.101
(1.912)*
Initial human capital
0.009
(1.161)
-0.01
(-0.923)
-0.01
(-1.685)*
-0.001
(-0.094)
Tax revenue
(% of GDP)
-0.216
(-1.883)*
0.036
(0.324)
-0.015
(-0.202)
-0.029
(-0.376)
Other revenue
(% of GDP)
-0.407
(-3.664)***
-0.287
(-1.13)
-0.153
(-1.944)*
-0.300
(-3.072)***
Productive + nonproductive expenditure (-1)
(% of GDP)
0.225
(0.486)
0.088
(0.552)
0.054
(0.456)
…
Square term of productive + nonproductive expenditure (-1)
(% of GDP)
0.673
(0.652)
-0.275
(-0.502)
0.018
(0.047)
…
DFAST*Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
0.046
(0.323)
DFAST*Square term of Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
0.548
(1.052)
DCOMP*Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
0.069
(0.568)
DCOMP*Square term of Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
-0.158
(-0.382)
Budget surplus
(% of GDP)
0.378
(3.524)***
0.241
(2.052)**
0.186
(2.557)**
0.252
(3.328)***
Inflation - consumer price index
-0.210
(-5.494)***
-0.01
(-0.803)
-0.023
(-2.326)**
-0.016
(-1.566)
Growth rate of GDP per capita (-1)
0.386
(3.699)***
0.268
(3.356)***
0.285
(4.467)***
0.279
(4.413)***
No. of observations
Adjusted R2
99
0.442
171
0.096
270
0.311
270
0.326
Note: The estimation method is the ordinary least squares for panel data. Annual data are used. ( -1) indicates variables lagged one period.
DFAST is the dummy variable which is 1 for fast growing countries and 0 otherwise. DCOMP is the dummy variable which is 1 for the
comparison group and 0 otherwise. t-statistics are given in parenthesis. * indicates 10% significance level, ** indicates 5% significance
level, and *** indicates 1% significance level. These significance levels are equal to one minus the probability of rejecting the null
hypothesis of zero coefficients.
33
Table 8: Nonlinear Results with Total Public Expenditures (Dynamic Panel - GMM)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
Comparison
group
Whole sample
pooled
Whole sample
with dummies
Openness
(% of GDP)
0.044
(2.106)**
-0.281
(-1.344)
0.068
(1.971)**
0.014
(0.171)
Total government revenue
(% of GDP)
-0.492
(-1.58)
0.192
(0.332)
-0.443
(-1.062)
-0.393
(-0.653)
Productive + nonproductive expenditure (-1)
(% of GDP)
0.508
(0.547)
0.775
(0.735)
1.098
(1.104)
…
Square term of productive + nonproductive expenditure (-1)
(% of GDP)
-0.997
(-0.465)
-3.210
(-1.004)
-3.408
(-1.349)
…
DFAST*Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
2.593
(0.63)
DFAST*Square term of Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
-2.972
(-0.339)
DCOMP*Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
2.943
(1.468)
DCOMP*Square term of Productive + nonproductive expenditure (-1)
(% of GDP)
…
…
…
-11.894
(-1.565)
Budget surplus
(% of GDP)
-0.089
(-0.261)
0.051
(0.081)
0.278
(0.563)
0.099
(0.157)
Inflation - consumer price index
-0.388
(-4.015)***
-0.052
(-2.672)***
-0.065
(-2.341)**
-0.055
(-1.521)
Growth rate of GDP per capita (-1)
0.082
(0.484)
-0.400
(-2.513)**
-0.650
(-4.742)***
-0.386
(-1.544)
No. of observations
J-test
99
1.035
171
2.514
270
1.346
270
2.996
Note: The estimation method is a dynamic GMM. Annual data are used. (-1) indicates variables lagged one period. DFAST is the dummy
variable which is 1 f or f ast growing countries and 0 otherwise. DCOMP is the dummy variable which is 1 f or the comparison group and 0
otherwise. t-statistics are given in parenthesis. * indicates 10% signif icance level, ** indicates 5% signif icance level, and *** indicate s 1%
signif icance level. These signif icance levels are equal to one minus the probability of rejecting the null hypothesis of zero coef f icients. J-test
is f or overidentif ication problem where H0: there is no overidentif ication problem. We f ail to reject in each case.
34
Table 9: Nonlinearity of Productive Expenditures (panel OLS)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
Comparison
group
Whole sample
pooled
Whole sample
with dummies
Private investment
(% of GDP)
0.072
(1.238)
0.096
(0.829)
0.161
(3.337)***
0.099
(1.807)*
Initial human capital
0.006
(0.791)
-0.015
(-1.363)
-0.009
(-1.514)
-0.004
(-0.584)
Tax revenue
(% of GDP)
-0.213
(-1.895)*
-0.011
(-0.084)
-0.037
(-0.451)
-0.050
(-0.597)
Other revenue
(% of GDP)
-0.540
(-4.039)***
-0.298
(-1.284)
-0.190
(-1.98)**
-0.303
(-2.915)***
Productive expenditure (-1)
(% of GDP)
0.826
(1.756)*
0.153
(0.545)
-0.080
(-0.473)
…
Square term of Productive expenditure (-1)
(% of GDP)
-0.464
(-0.365)
-1.133
(-0.764)
0.621
(0.823)
…
DFAST*Productive expenditure (-1)
(% of GDP)
…
…
…
0.093
(0.507)
DFAST*Square term of Productive expenditure (-1)
(% of GDP)
…
…
…
0.611
(0.793)
DCOMP*Productive expenditure (-1)
(% of GDP)
…
…
…
0.137
(0.613)
DCOMP*Square term of Productive expenditure (-1) …
(% of GDP)
…
…
-0.943
(-0.77)
Non-productive expenditure (-1)
(% of GDP)
0.316
(1.799)*
0.068
(0.552)
0.097
(1.1)
0.077
(0.859)
Budget surplus
(% of GDP)
0.463
(4.073)***
0.264
(2.207)**
0.194
(2.67)***
0.249
(3.32)***
Inflation - consumer price index
-0.205
(-5.456)***
-0.012
(-0.959)
-0.024
(-2.422)**
-0.018
(-1.769)*
Growth rate of GDP per capita (-1)
0.376
(3.617)***
0.259
(3.216)***
0.284
(4.436)***
0.271
(4.263)***
No. of observations
Adjusted R2
120
0.449
167
0.097
287
0.311
287
0.326
Note: The estimation method is the ordinary least squares f or panel data. Annual data are used. (-1) indicates variables
lagged one period. DFAST is the dummy variable which is 1 f or f ast growing countries and 0 otherwise. DCOMP is the
dummy variable which is 1 f or the comparison group and 0 otherwise. t-statistics are given in parenthesis. * indicates 10%
signif icance level, ** indicates 5% signif icance level, and *** indicates 1% signif icance level. These signif icance levels ar e
equal to one minus the probability of rejecting the null hypothesis of zero coef f icients.
35
Table 10: Nonlinear Results with Productive and Non-productive Expenditures
(Dynamic Panel - GMM)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
Comparison
group
Whole sample
pooled
Whole sample
with dummies
Openness
(% of GDP)
0.024
(0.734)
-0.197
(-0.96)
0.056
(1.026)
0.014
(0.118)
Total government revenue
(% of GDP)
-0.665
(-2.164)**
0.655
(1.102)
0.050
(0.113)
0.017
(0.03)
Productive expenditure (-1)
(% of GDP)
6.061
(1.687)*
0.659
(0.456)
1.06
(1.037)
…
Square term of Productive expenditure (-1)
(% of GDP)
-12.969
(-1.508)
0.354
(0.046)
-0.687
(-0.187)
…
DFAST*Productive expenditure (-1)
(% of GDP)
…
…
…
4.016
(0.457)
DFAST*Square term of Productive expenditure (-1)
(% of GDP)
…
…
…
-8.252
(-0.426)
DCOMP*Productive expenditure (-1)
(% of GDP)
…
…
…
0.209
(0.089)
DCOMP*Square term of Productive expenditure (-1)
(% of GDP)
…
…
…
3.393
(0.212)
Non-productive expenditure (-1)
(% of GDP)
-2.125
(-2.5)**
-1.179
(-1.608)
-1.377
(-2.06)**
-1.440
(-2.031)**
Budget surplus
(% of GDP)
0.735
(1.58)
-0.225
(-0.321)
-0.009
(-0.024)
0.073
(0.114)
Inflation - consumer price index
-0.214
(-1.816)*
-0.089
(-3.245)***
-0.08
(-4.407)***
-0.088
(-1.866)*
Growth rate of GDP per capita (-1)
-0.306
(-1.487)
-0.387
(-2.328)**
-0.393
(-2.564)**
-0.355
(-1.553)
No. of observations
J-test
99
1.007
171
2.321
270
2.511
270
2.919
Note: The estimation method is a dynamic GMM. Annual data are used. (-1) indicates variables lagged one period. DFAST is
the dummy variable which is 1 f or f ast growing countries and 0 otherwise. DCOMP is the dummy variable which is 1 f or the
comparison group and 0 otherwise. t-statistics are given in parenthesis. * indicates 10% signif icance level, ** indicates 5%
signif icance level, and *** indicates 1% signif icance level. These signif icance levels are equal to one minus the probability of
rejecting the null hypothesis of zero coef f icients. J -test is f or overidentification problem where H0: there is no
overidentif ication problem. We f ail to reject in each case.
36
Table 11: Results with Alternative Definiiton of Productive Expenditures (panel OLS)
Dependent variable: Growth rate of GDP per capita
Fast growing
countries
(1)
Comparison
group
(2)
Whole sample
pooled
(3)
Whole sample
with dummies
(4)
Private investment
(% of GDP)
0.095
(1.597)
0.116
(1.021)
0.163
(3.525)***
0.104
(2.016)**
Initial human capital
0.009
(1.091)
-0.016
(-1.402)
-0.011
(-1.795)*
-0.003
(-0.472)
Tax revenue
(% of GDP)
-0.168
(-1.564)
-0.033
(-0.274)
-0.032
(-0.398)
-0.044
(-0.527)
Other revenue
(% of GDP)
-0.264
(-2.173)**
-0.364
(-1.61)
-0.105
(-1.429)
-0.236
(-2.99)***
Productive expenditure (-1)
(% of GDP)
0.400
(2.348)**
-0.074
(-0.754)
-0.005
(-0.071)
…
DFAST*Productive expenditure (-1)
(% of GDP)
…
…
…
0.259
(2.7)***
DCOMP*Productive expenditure (-1) …
(% of GDP)
…
…
0.158
(1.311)
Non-productive expenditure (-1)
(% of GDP)
0.029
(0.153)
0.106
(0.961)
0.098
(1.175)
-0.075
(-0.792)
Budget surplus
(% of GDP)
0.276
(2.621)***
0.274
(2.315)**
0.165
(2.347)**
0.223
(2.991)***
Inflation - consumer price index
-0.210
(-5.545)***
-0.014
(-1.073)
-0.024
(-2.492)**
-0.020
(-1.934)*
Growth rate of GDP per capita (-1)
0.371
(3.422)***
0.252
(3.14)***
0.280
(4.373)***
0.270
(4.271)***
No. of observations
Adjusted R2
122
0.390
167
0.102
289
0.312
289
0.324
Note: The alternative def inition of the productive component is the original def inition of productive spending minus
def ense spending. The estimation method is the ordinary least squares f or panel data. Annual data are used. (-1)
indicates variables lagged one period. D-FAST is the dummy variable which is 1 f or f ast growing countries and 0
otherwise. DCOMP is the dummy variable which is 1 f or the comparison group and 0 otherwise. t-statistics are given
in parenthesis. * indicates 10% signif icance level, ** indicates 5% signif icance level, and *** indicates 1%
signif icance level. These signif icance levels are equal to one minus the probability of rejecting the null hypothesis of
zero coef f icients.
37
Da
Sa
FIGURES
Figure 1 - GDP per capita growth (annual %)
12
10
8
6
4
2
0
-2
-4
-6
-8
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
Fast growing countries
Other developing countries
Figure 2 - GDP (in constant 2000 US$) per
labor force
18000
16000
14000
12000
10000
8000
6000
4000
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
Fast growing countries
Other developing countries
38
Figure 3 - Industry, value added (% of GDP)
45
40
35
30
25
20
2004
2006
2002
2004
2002
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
Fast growing countries
Other developing countries
Figure 4 - Productive component of public
spending (in % of total expenditure)
90
80
70
60
50
40
30
2000
1998
1996
1994
1992
1990
1988
1986
1984
1982
1980
1978
1976
1974
1972
1970
Fast growing countries
Other developing countries
39
Figure 5 - Gross public fixed capital formation
(in % of GDP)
14
12
10
8
6
4
2
0
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
Venezuela
Uruguay
Turkey
Philippines
Mexico
Costa Rica
Chile
Thailand
Singapore
Mauritius
Malaysia
Korea, Rep.
Indonesia
Botswana
40
2007
1996
1993
1992
1991
1990
1989
1988
1987
1986
1985
1984
1983
1982
1981
1980
Other developing countries
Fast growing countries
Figure 6 - Government Effectiveness
100
90
80
70
60
50
40
30
20
10
0
Figure 7 - Average Bureaucracy Quality
(index between 0-4, higher index higher quality)
3.1
2.9
2.7
2.5
2.3
2.1
1.9
1.7
1.5
2008
2007
41
2006
Other developing countries
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
1991
1990
1989
1988
1987
1986
1985
Fast growing countries